A comparison of neural network-based super-resolution models on 3D rendered images.

In this project we compare three different approaches based on some state-of-the-art neural-network based super-resolution techniques from both the perspective of quality and computational cost, when used to enhance 3D images obtained from video games.

Abstract

Super-resolution is an area of Computer Vision comprising various techniques to recover a high-resolution image from a low-resolution counterpart. These techniques can also be used to enhance a low-resolution input image without a native high-resolution original. Single Image Super-Resolution (SISR) techniques aim to do this in a picture-by-picture fashion. In recent years, deep learning models have achieved the best performance, using neural networks to find a mapping between an input low-resolution image and its high-resolution analogous. This work will compare three approaches based on some of the most notable works in neural-network based super-resolution: SRCNN, FSRCNN, and ESRGAN. These methods will be used to enhance 3D computer-generated low-resolution pictures obtained from popular video games and will be evaluated with respect to the quality of the enhanced picture and the required computation time. From our study, we can attest to the superiority of neural network-based methods on the SISR problem and the benefits of taking a perceptual approach when the quality of the resulting images.

Code

To support the research community and encourage exploration, we have provided access to the code used for the development and testing of our models through our GitHub repository. Also, you can easily try our methods with no installation required through our Google Colab demo.

Citing

If you use this work in your research, you must cite:

  1. Rafael Berral-Soler, Francisco J. Madrid-Cuevas, Sebastián Ventura, Rafael Muñoz-Salinas, and Manuel J. Marín-Jiménez. 2023. A Comparison of Neural Network-Based Super-Resolution Models on 3D Rendered Images. In Computer Analysis of Images and Patterns: 20th International Conference, CAIP 2023, Limassol, Cyprus, September 25–28, 2023, Proceedings, Part I. Springer-Verlag, Berlin, Heidelberg, 45–55. https://doi.org/10.1007/978-3-031-44237-7_5

Contact

If you have any further questions, please contact rberral@uco.es.